Yue Cao

Yuxin Fang, Wen Wang, Binhui Xie et al. · meta-ai

We launch EVA, a vision-centric foundation model to explore the limits of visual representation at scale using only publicly accessible data. EVA is a vanilla ViT pre-trained to reconstruct the masked out image-text aligned vision features conditioned on visible image patches. Via this pretext task, we can efficiently scale up EVA to one billion parameters, and sets new records on a broad range of representative vision downstream tasks, such as image recognition, video action recognition, object detection, instance segmentation and semantic segmentation without heavy supervised training. Moreover, we observe quantitative changes in scaling EVA result in qualitative changes in transfer learning performance that are not present in other models. For instance, EVA takes a great leap in the challenging large vocabulary instance segmentation task: our model achieves almost the same state-of-the-art performance on LVISv1.0 dataset with over a thousand categories and COCO dataset with only eighty categories. Beyond a pure vision encoder, EVA can also serve as a vision-centric, multi-modal pivot to connect images and text. We find initializing the vision tower of a giant CLIP from EVA can greatly stabilize the training and outperform the training from scratch counterpart with much fewer samples and less compute, providing a new direction for scaling up and accelerating the costly training of multi-modal foundation models. To facilitate future research, we release all the code and models at https://github.com/baaivision/EVA.

Quan Sun, Yuxin Fang, Ledell Wu et al. · meta-ai

Contrastive language-image pre-training, CLIP for short, has gained increasing attention for its potential in various scenarios. In this paper, we propose EVA-CLIP, a series of models that significantly improve the efficiency and effectiveness of CLIP training. Our approach incorporates new techniques for representation learning, optimization, and augmentation, enabling EVA-CLIP to achieve superior performance compared to previous CLIP models with the same number of parameters but significantly smaller training costs. Notably, our largest 5.0B-parameter EVA-02-CLIP-E/14+ with only 9 billion seen samples achieves 82.0 zero-shot top-1 accuracy on ImageNet-1K val. A smaller EVA-02-CLIP-L/14+ with only 430 million parameters and 6 billion seen samples achieves 80.4 zero-shot top-1 accuracy on ImageNet-1K val. To facilitate open access and open research, we release the complete suite of EVA-CLIP to the community at https://github.com/baaivision/EVA/tree/master/EVA-CLIP.

Zanlin Ni, Yulin Wang, Jiangwei Yu et al. · tsinghua

Recent years have witnessed a remarkable success of large deep learning models. However, training these models is challenging due to high computational costs, painfully slow convergence, and overfitting issues. In this paper, we present Deep Incubation, a novel approach that enables the efficient and effective training of large models by dividing them into smaller sub-modules that can be trained separately and assembled seamlessly. A key challenge for implementing this idea is to ensure the compatibility of the independently trained sub-modules. To address this issue, we first introduce a global, shared meta model, which is leveraged to implicitly link all the modules together, and can be designed as an extremely small network with negligible computational overhead. Then we propose a module incubation algorithm, which trains each sub-module to replace the corresponding component of the meta model and accomplish a given learning task. Despite the simplicity, our approach effectively encourages each sub-module to be aware of its role in the target large model, such that the finally-learned sub-modules can collaborate with each other smoothly after being assembled. Empirically, our method outperforms end-to-end (E2E) training in terms of both final accuracy and training efficiency. For example, on top of ViT-Huge, it improves the accuracy by 2.7% on ImageNet or achieves similar performance with 4x less training time. Notably, the gains are significant for downstream tasks as well (e.g., object detection and image segmentation on COCO and ADE20K). Code is available at https://github.com/LeapLabTHU/Deep-Incubation.

Yuxin Fang, Quan Sun, Xinggang Wang et al.

We launch EVA-02, a next-generation Transformer-based visual representation pre-trained to reconstruct strong and robust language-aligned vision features via masked image modeling. With an updated plain Transformer architecture as well as extensive pre-training from an open & accessible giant CLIP vision encoder, EVA-02 demonstrates superior performance compared to prior state-of-the-art approaches across various representative vision tasks, while utilizing significantly fewer parameters and compute budgets. Notably, using exclusively publicly accessible training data, EVA-02 with only 304M parameters achieves a phenomenal 90.0 fine-tuning top-1 accuracy on ImageNet-1K val set. Additionally, our EVA-02-CLIP can reach up to 80.4 zero-shot top-1 on ImageNet-1K, outperforming the previous largest & best open-sourced CLIP with only ~1/6 parameters and ~1/6 image-text training data. We offer four EVA-02 variants in various model sizes, ranging from 6M to 304M parameters, all with impressive performance. To facilitate open access and open research, we release the complete suite of EVA-02 to the community at https://github.com/baaivision/EVA/tree/master/EVA-02.

Wen Wang, Yan Jiang, Kangyang Xie et al.

Large-scale text-to-image diffusion models achieve unprecedented success in image generation and editing. However, how to extend such success to video editing is unclear. Recent initial attempts at video editing require significant text-to-video data and computation resources for training, which is often not accessible. In this work, we propose vid2vid-zero, a simple yet effective method for zero-shot video editing. Our vid2vid-zero leverages off-the-shelf image diffusion models, and doesn't require training on any video. At the core of our method is a null-text inversion module for text-to-video alignment, a cross-frame modeling module for temporal consistency, and a spatial regularization module for fidelity to the original video. Without any training, we leverage the dynamic nature of the attention mechanism to enable bi-directional temporal modeling at test time. Experiments and analyses show promising results in editing attributes, subjects, places, etc., in real-world videos. Code is made available at \url{https://github.com/baaivision/vid2vid-zero}.

Yixuan Wei, Han Hu, Zhenda Xie et al.

Masked image modeling (MIM) learns representations with remarkably good fine-tuning performances, overshadowing previous prevalent pre-training approaches such as image classification, instance contrastive learning, and image-text alignment. In this paper, we show that the inferior fine-tuning performance of these pre-training approaches can be significantly improved by a simple post-processing in the form of feature distillation (FD). The feature distillation converts the old representations to new representations that have a few desirable properties just like those representations produced by MIM. These properties, which we aggregately refer to as optimization friendliness, are identified and analyzed by a set of attention- and optimization-related diagnosis tools. With these properties, the new representations show strong fine-tuning performance. Specifically, the contrastive self-supervised learning methods are made as competitive in fine-tuning as the state-of-the-art masked image modeling (MIM) algorithms. The CLIP models' fine-tuning performance is also significantly improved, with a CLIP ViT-L model reaching 89.0% top-1 accuracy on ImageNet-1K classification. On the 3-billion-parameter SwinV2-G model, the fine-tuning accuracy is improved by +1.5 mIoU / +1.1 mAP to 61.4 mIoU / 64.2 mAP on ADE20K semantic segmentation and COCO object detection, respectively, creating new records on both benchmarks. More importantly, our work provides a way for the future research to focus more effort on the generality and scalability of the learnt representations without being pre-occupied with optimization friendliness since it can be enhanced rather easily. The code will be available at https://github.com/SwinTransformer/Feature-Distillation.

Yixuan Wei, Yue Cao, Zheng Zhang et al.

Image classification, which classifies images by pre-defined categories, has been the dominant approach to visual representation learning over the last decade. Visual learning through image-text alignment, however, has emerged to show promising performance, especially for zero-shot recognition. We believe that these two learning tasks are complementary, and suggest combining them for better visual learning. We propose a deep fusion method with three adaptations that effectively bridge two learning tasks, rather than shallow fusion through naive multi-task learning. First, we modify the previous common practice in image classification, a linear classifier, with a cosine classifier which shows comparable performance. Second, we convert the image classification problem from learning parametric category classifier weights to learning a text encoder as a meta network to generate category classifier weights. The learnt text encoder is shared between image classification and image-text alignment. Third, we enrich each class name with a description to avoid confusion between classes and make the classification method closer to the image-text alignment. We prove that this deep fusion approach performs better on a variety of visual recognition tasks and setups than the individual learning or shallow fusion approach, from zero-shot/few-shot image classification, such as the Kornblith 12-dataset benchmark, to downstream tasks of action recognition, semantic segmentation, and object detection in fine-tuning and open-vocabulary settings. The code will be available at https://github.com/weiyx16/iCAR.

Xinlong Wang, Wen Wang, Yue Cao et al.

In-context learning, as a new paradigm in NLP, allows the model to rapidly adapt to various tasks with only a handful of prompts and examples. But in computer vision, the difficulties for in-context learning lie in that tasks vary significantly in the output representations, thus it is unclear how to define the general-purpose task prompts that the vision model can understand and transfer to out-of-domain tasks. In this work, we present Painter, a generalist model which addresses these obstacles with an "image"-centric solution, that is, to redefine the output of core vision tasks as images, and specify task prompts as also images. With this idea, our training process is extremely simple, which performs standard masked image modeling on the stitch of input and output image pairs. This makes the model capable of performing tasks conditioned on visible image patches. Thus, during inference, we can adopt a pair of input and output images from the same task as the input condition, to indicate which task to perform. Without bells and whistles, our generalist Painter can achieve competitive performance compared to well-established task-specific models, on seven representative vision tasks ranging from high-level visual understanding to low-level image processing. In addition, Painter significantly outperforms recent generalist models on several challenging tasks.

Chenyu Zheng, Guoqiang Wu, Fan Bao et al.

A large-scale deep model pre-trained on massive labeled or unlabeled data transfers well to downstream tasks. Linear evaluation freezes parameters in the pre-trained model and trains a linear classifier separately, which is efficient and attractive for transfer. However, little work has investigated the classifier in linear evaluation except for the default logistic regression. Inspired by the statistical efficiency of naive Bayes, the paper revisits the classical topic on discriminative vs. generative classifiers. Theoretically, the paper considers the surrogate loss instead of the zero-one loss in analyses and generalizes the classical results from binary cases to multiclass ones. We show that, under mild assumptions, multiclass naive Bayes requires $O(\log n)$ samples to approach its asymptotic error while the corresponding multiclass logistic regression requires $O(n)$ samples, where $n$ is the feature dimension. To establish it, we present a multiclass $\mathcal{H}$-consistency bound framework and an explicit bound for logistic loss, which are of independent interests. Simulation results on a mixture of Gaussian validate our theoretical findings. Experiments on various pre-trained deep vision models show that naive Bayes consistently converges faster as the number of data increases. Besides, naive Bayes shows promise in few-shot cases and we observe the "two regimes" phenomenon in pre-trained supervised models. Our code is available at https://github.com/ML-GSAI/Revisiting-Dis-vs-Gen-Classifiers.

Xinlong Wang, Xiaosong Zhang, Yue Cao et al.

We present SegGPT, a generalist model for segmenting everything in context. We unify various segmentation tasks into a generalist in-context learning framework that accommodates different kinds of segmentation data by transforming them into the same format of images. The training of SegGPT is formulated as an in-context coloring problem with random color mapping for each data sample. The objective is to accomplish diverse tasks according to the context, rather than relying on specific colors. After training, SegGPT can perform arbitrary segmentation tasks in images or videos via in-context inference, such as object instance, stuff, part, contour, and text. SegGPT is evaluated on a broad range of tasks, including few-shot semantic segmentation, video object segmentation, semantic segmentation, and panoptic segmentation. Our results show strong capabilities in segmenting in-domain and out-of-domain targets, either qualitatively or quantitatively.

Fan Bao, Shen Nie, Kaiwen Xue et al.

Vision transformers (ViT) have shown promise in various vision tasks while the U-Net based on a convolutional neural network (CNN) remains dominant in diffusion models. We design a simple and general ViT-based architecture (named U-ViT) for image generation with diffusion models. U-ViT is characterized by treating all inputs including the time, condition and noisy image patches as tokens and employing long skip connections between shallow and deep layers. We evaluate U-ViT in unconditional and class-conditional image generation, as well as text-to-image generation tasks, where U-ViT is comparable if not superior to a CNN-based U-Net of a similar size. In particular, latent diffusion models with U-ViT achieve record-breaking FID scores of 2.29 in class-conditional image generation on ImageNet 256x256, and 5.48 in text-to-image generation on MS-COCO, among methods without accessing large external datasets during the training of generative models. Our results suggest that, for diffusion-based image modeling, the long skip connection is crucial while the down-sampling and up-sampling operators in CNN-based U-Net are not always necessary. We believe that U-ViT can provide insights for future research on backbones in diffusion models and benefit generative modeling on large scale cross-modality datasets.

Fei Xie, Chunyu Wang, Guangting Wang et al.

Robustness and discrimination power are two fundamental requirements in visual object tracking. In most tracking paradigms, we find that the features extracted by the popular Siamese-like networks cannot fully discriminatively model the tracked targets and distractor objects, hindering them from simultaneously meeting these two requirements. While most methods focus on designing robust correlation operations, we propose a novel target-dependent feature network inspired by the self-/cross-attention scheme. In contrast to the Siamese-like feature extraction, our network deeply embeds cross-image feature correlation in multiple layers of the feature network. By extensively matching the features of the two images through multiple layers, it is able to suppress non-target features, resulting in instance-varying feature extraction. The output features of the search image can be directly used for predicting target locations without extra correlation step. Moreover, our model can be flexibly pre-trained on abundant unpaired images, leading to notably faster convergence than the existing methods. Extensive experiments show our method achieves the state-of-the-art results while running at real-time. Our feature networks also can be applied to existing tracking pipelines seamlessly to raise the tracking performance. Code will be available.

Fan Bao, Shen Nie, Kaiwen Xue et al.

This paper proposes a unified diffusion framework (dubbed UniDiffuser) to fit all distributions relevant to a set of multi-modal data in one model. Our key insight is -- learning diffusion models for marginal, conditional, and joint distributions can be unified as predicting the noise in the perturbed data, where the perturbation levels (i.e. timesteps) can be different for different modalities. Inspired by the unified view, UniDiffuser learns all distributions simultaneously with a minimal modification to the original diffusion model -- perturbs data in all modalities instead of a single modality, inputs individual timesteps in different modalities, and predicts the noise of all modalities instead of a single modality. UniDiffuser is parameterized by a transformer for diffusion models to handle input types of different modalities. Implemented on large-scale paired image-text data, UniDiffuser is able to perform image, text, text-to-image, image-to-text, and image-text pair generation by setting proper timesteps without additional overhead. In particular, UniDiffuser is able to produce perceptually realistic samples in all tasks and its quantitative results (e.g., the FID and CLIP score) are not only superior to existing general-purpose models but also comparable to the bespoken models (e.g., Stable Diffusion and DALL-E 2) in representative tasks (e.g., text-to-image generation).

Yue Cao, Zhaolin Wan, Dongwei Ren et al.

Full-reference (FR) image quality assessment (IQA) evaluates the visual quality of a distorted image by measuring its perceptual difference with pristine-quality reference, and has been widely used in low-level vision tasks. Pairwise labeled data with mean opinion score (MOS) are required in training FR-IQA model, but is time-consuming and cumbersome to collect. In contrast, unlabeled data can be easily collected from an image degradation or restoration process, making it encouraging to exploit unlabeled training data to boost FR-IQA performance. Moreover, due to the distribution inconsistency between labeled and unlabeled data, outliers may occur in unlabeled data, further increasing the training difficulty. In this paper, we suggest to incorporate semi-supervised and positive-unlabeled (PU) learning for exploiting unlabeled data while mitigating the adverse effect of outliers. Particularly, by treating all labeled data as positive samples, PU learning is leveraged to identify negative samples (i.e., outliers) from unlabeled data. Semi-supervised learning (SSL) is further deployed to exploit positive unlabeled data by dynamically generating pseudo-MOS. We adopt a dual-branch network including reference and distortion branches. Furthermore, spatial attention is introduced in the reference branch to concentrate more on the informative regions, and sliced Wasserstein distance is used for robust difference map computation to address the misalignment issues caused by images recovered by GAN models. Extensive experiments show that our method performs favorably against state-of-the-arts on the benchmark datasets PIPAL, KADID-10k, TID2013, LIVE and CSIQ.

Zhenda Xie, Zigang Geng, Jingcheng Hu et al.

Masked image modeling (MIM) as pre-training is shown to be effective for numerous vision downstream tasks, but how and where MIM works remain unclear. In this paper, we compare MIM with the long-dominant supervised pre-trained models from two perspectives, the visualizations and the experiments, to uncover their key representational differences. From the visualizations, we find that MIM brings locality inductive bias to all layers of the trained models, but supervised models tend to focus locally at lower layers but more globally at higher layers. That may be the reason why MIM helps Vision Transformers that have a very large receptive field to optimize. Using MIM, the model can maintain a large diversity on attention heads in all layers. But for supervised models, the diversity on attention heads almost disappears from the last three layers and less diversity harms the fine-tuning performance. From the experiments, we find that MIM models can perform significantly better on geometric and motion tasks with weak semantics or fine-grained classification tasks, than their supervised counterparts. Without bells and whistles, a standard MIM pre-trained SwinV2-L could achieve state-of-the-art performance on pose estimation (78.9 AP on COCO test-dev and 78.0 AP on CrowdPose), depth estimation (0.287 RMSE on NYUv2 and 1.966 RMSE on KITTI), and video object tracking (70.7 SUC on LaSOT). For the semantic understanding datasets where the categories are sufficiently covered by the supervised pre-training, MIM models can still achieve highly competitive transfer performance. With a deeper understanding of MIM, we hope that our work can inspire new and solid research in this direction.

Zhenda Xie, Zheng Zhang, Yue Cao et al.

An important goal of self-supervised learning is to enable model pre-training to benefit from almost unlimited data. However, one method that has recently become popular, namely masked image modeling (MIM), is suspected to be unable to benefit from larger data. In this work, we break this misconception through extensive experiments, with data scales ranging from 10\% of ImageNet-1K to full ImageNet-22K, model sizes ranging from 49 million to 1 billion, and training lengths ranging from 125K iterations to 500K iterations. Our study reveals that: (i) Masked image modeling is also demanding on larger data. We observed that very large models got over-fitted with relatively small data; (ii) The length of training matters. Large models trained with masked image modeling can benefit from more data with longer training; (iii) The validation loss in pre-training is a good indicator to measure how well the model performs for fine-tuning on multiple tasks. This observation allows us to pre-evaluate pre-trained models in advance without having to make costly trial-and-error assessments of downstream tasks. We hope that our findings will advance the understanding of masked image modeling in terms of scaling ability.

Yue Cao, Tianlin Li, Xiaofeng Cao et al.

We introduce a novel approach to counter adversarial attacks, namely, image resampling. Image resampling transforms a discrete image into a new one, simulating the process of scene recapturing or rerendering as specified by a geometrical transformation. The underlying rationale behind our idea is that image resampling can alleviate the influence of adversarial perturbations while preserving essential semantic information, thereby conferring an inherent advantage in defending against adversarial attacks. To validate this concept, we present a comprehensive study on leveraging image resampling to defend against adversarial attacks. We have developed basic resampling methods that employ interpolation strategies and coordinate shifting magnitudes. Our analysis reveals that these basic methods can partially mitigate adversarial attacks. However, they come with apparent limitations: the accuracy of clean images noticeably decreases, while the improvement in accuracy on adversarial examples is not substantial. We propose implicit representation-driven image resampling (IRAD) to overcome these limitations. First, we construct an implicit continuous representation that enables us to represent any input image within a continuous coordinate space. Second, we introduce SampleNet, which automatically generates pixel-wise shifts for resampling in response to different inputs. Furthermore, we can extend our approach to the state-of-the-art diffusion-based method, accelerating it with fewer time steps while preserving its defense capability. Extensive experiments demonstrate that our method significantly enhances the adversarial robustness of diverse deep models against various attacks while maintaining high accuracy on clean images.

Yutong Lin, Ze Liu, Zheng Zhang et al.

Frozen pretrained models have become a viable alternative to the pretraining-then-finetuning paradigm for transfer learning. However, with frozen models there are relatively few parameters available for adapting to downstream tasks, which is problematic in computer vision where tasks vary significantly in input/output format and the type of information that is of value. In this paper, we present a study of frozen pretrained models when applied to diverse and representative computer vision tasks, including object detection, semantic segmentation and video action recognition. From this empirical analysis, our work answers the questions of what pretraining task fits best with this frozen setting, how to make the frozen setting more flexible to various downstream tasks, and the effect of larger model sizes. We additionally examine the upper bound of performance using a giant frozen pretrained model with 3 billion parameters (SwinV2-G) and find that it reaches competitive performance on a varied set of major benchmarks with only one shared frozen base network: 60.0 box mAP and 52.2 mask mAP on COCO object detection test-dev, 57.6 val mIoU on ADE20K semantic segmentation, and 81.7 top-1 accuracy on Kinetics-400 action recognition. With this work, we hope to bring greater attention to this promising path of freezing pretrained image models.

Weiyun Wang, Zhe Chen, Wenhai Wang et al.

Existing open-source multimodal large language models (MLLMs) generally follow a training process involving pre-training and supervised fine-tuning. However, these models suffer from distribution shifts, which limit their multimodal reasoning, particularly in the Chain-of-Thought (CoT) performance. To address this, we introduce a preference optimization (PO) process to enhance the multimodal reasoning capabilities of MLLMs. Specifically, (1) on the data side, we design an automated preference data construction pipeline to create MMPR, a high-quality, large-scale multimodal reasoning preference dataset; and (2) on the model side, we explore integrating PO with MLLMs, developing a simple yet effective method, termed Mixed Preference Optimization (MPO), which boosts multimodal CoT performance. Our approach enhances the multimodal reasoning abilities of both InternVL2-8B and InternVL2-76B. Notably, our model, InternVL2-8B-MPO, achieves an accuracy of 67.0 on MathVista, outperforming InternVL2-8B by 8.7 points and achieving performance comparable to the 10$\times$ larger InternVL2-76B. We hope this study could inspire further advancements in MLLMs. Code, data, and model are released.

Yue Cao, XiaoJiang Zhou, Jiaqi Feng et al.

Rich user behavior data has been proven to be of great value for Click-Through Rate (CTR) prediction applications, especially in industrial recommender, search, or advertising systems. However, it's non-trivial for real-world systems to make full use of long-term user behaviors due to the strict requirements of online serving time. Most previous works adopt the retrieval-based strategy, where a small number of user behaviors are retrieved first for subsequent attention. However, the retrieval-based methods are sub-optimal and would cause more or less information losses, and it's difficult to balance the effectiveness and efficiency of the retrieval algorithm. In this paper, we propose SDIM (Sampling-based Deep Interest Modeling), a simple yet effective sampling-based end-to-end approach for modeling long-term user behaviors. We sample from multiple hash functions to generate hash signatures of the candidate item and each item in the user behavior sequence, and obtain the user interest by directly gathering behavior items associated with the candidate item with the same hash signature. We show theoretically and experimentally that the proposed method performs on par with standard attention-based models on modeling long-term user behaviors, while being sizable times faster. We also introduce the deployment of SDIM in our system. Specifically, we decouple the behavior sequence hashing, which is the most time-consuming part, from the CTR model by designing a separate module named BSE (behavior Sequence Encoding). BSE is latency-free for the CTR server, enabling us to model extremely long user behaviors. Both offline and online experiments are conducted to demonstrate the effectiveness of SDIM. SDIM now has been deployed online in the search system of Meituan APP.

Yue Cao, C. S. George Lee

Nowadays, the behavior tree is gaining popularity as a representation for robot tasks due to its modularity and reusability. Designing behavior-tree tasks manually is time-consuming for robot end-users, thus there is a need for investigating automatic behavior-tree-based task generation. Prior behavior-tree-based task generation approaches focus on fixed primitive tasks and lack generalizability to new task domains. To cope with this issue, we propose a novel behavior-tree-based task generation approach that utilizes state-of-the-art large language models. We propose a Phase-Step prompt design that enables a hierarchical-structured robot task generation and further integrate it with behavior-tree-embedding-based search to set up the appropriate prompt. In this way, we enable an automatic and cross-domain behavior-tree task generation. Our behavior-tree-based task generation approach does not require a set of pre-defined primitive tasks. End-users only need to describe an abstract desired task and our proposed approach can swiftly generate the corresponding behavior tree. A full-process case study is provided to demonstrate our proposed approach. An ablation study is conducted to evaluate the effectiveness of our Phase-Step prompts. Assessment on Phase-Step prompts and the limitation of large language models are presented and discussed.

Yihong Chen, Yue Cao, Han Hu et al.

How do humans recognize an object in a piece of video? Due to the deteriorated quality of single frame, it may be hard for people to identify an occluded object in this frame by just utilizing information within one image. We argue that there are two important cues for humans to recognize objects in videos: the global semantic information and the local localization information. Recently, plenty of methods adopt the self-attention mechanisms to enhance the features in key frame with either global semantic information or local localization information. In this paper we introduce memory enhanced global-local aggregation (MEGA) network, which is among the first trials that takes full consideration of both global and local information. Furthermore, empowered by a novel and carefully-designed Long Range Memory (LRM) module, our proposed MEGA could enable the key frame to get access to much more content than any previous methods. Enhanced by these two sources of information, our method achieves state-of-the-art performance on ImageNet VID dataset. Code is available at \url{https://github.com/Scalsol/mega.pytorch}.

Weijie Su, Xizhou Zhu, Yue Cao et al.

We introduce a new pre-trainable generic representation for visual-linguistic tasks, called Visual-Linguistic BERT (VL-BERT for short). VL-BERT adopts the simple yet powerful Transformer model as the backbone, and extends it to take both visual and linguistic embedded features as input. In it, each element of the input is either of a word from the input sentence, or a region-of-interest (RoI) from the input image. It is designed to fit for most of the visual-linguistic downstream tasks. To better exploit the generic representation, we pre-train VL-BERT on the massive-scale Conceptual Captions dataset, together with text-only corpus. Extensive empirical analysis demonstrates that the pre-training procedure can better align the visual-linguistic clues and benefit the downstream tasks, such as visual commonsense reasoning, visual question answering and referring expression comprehension. It is worth noting that VL-BERT achieved the first place of single model on the leaderboard of the VCR benchmark. Code is released at \url{https://github.com/jackroos/VL-BERT}.

Yue Cao, Yang Shen

Motivation: Ab initio protein docking represents a major challenge for optimizing a noisy and costly "black box"-like function in a high-dimensional space. Despite progress in this field, there is no docking method available for rigorous uncertainty quantification (UQ) of its solution quality (e.g. interface RMSD or iRMSD). Results: We introduce a novel algorithm, Bayesian Active Learning (BAL), for optimization and UQ of such black-box functions and flexible protein docking. BAL directly models the posterior distribution of the global optimum (or native structures for protein docking) with active sampling and posterior estimation iteratively feeding each other. Furthermore, we use complex normal modes to represent a homogeneous Euclidean conformation space suitable for high-dimension optimization and construct funnel-like energy models for encounter complexes. Over a protein docking benchmark set and a CAPRI set including homology docking, we establish that BAL significantly improve against both starting points by rigid docking and refinements by particle swarm optimization, providing for one third targets a top-3 near-native prediction. BAL also generates tight confidence intervals with half range around 25% of iRMSD and confidence level at 85%. Its estimated probability of a prediction being native or not achieves binary classification AUROC at 0.93 and AUPRC over 0.60 (compared to 0.14 by chance); and also found to help ranking predictions. To the best of our knowledge, this study represents the first uncertainty quantification solution for protein docking, with theoretical rigor and comprehensive assessment. Source codes are available at https://github.com/Shen-Lab/BAL.

Fei Xie, Wankou Yang, Chunyu Wang et al.

Developing robust and discriminative appearance models has been a long-standing research challenge in visual object tracking. In the prevalent Siamese-based paradigm, the features extracted by the Siamese-like networks are often insufficient to model the tracked targets and distractor objects, thereby hindering them from being robust and discriminative simultaneously. While most Siamese trackers focus on designing robust correlation operations, we propose a novel single-branch tracking framework inspired by the transformer. Unlike the Siamese-like feature extraction, our tracker deeply embeds cross-image feature correlation in multiple layers of the feature network. By extensively matching the features of the two images through multiple layers, it can suppress non-target features, resulting in target-aware feature extraction. The output features can be directly used for predicting target locations without additional correlation steps. Thus, we reformulate the two-branch Siamese tracking as a conceptually simple, fully transformer-based Single-Branch Tracking pipeline, dubbed SBT. After conducting an in-depth analysis of the SBT baseline, we summarize many effective design principles and propose an improved tracker dubbed SuperSBT. SuperSBT adopts a hierarchical architecture with a local modeling layer to enhance shallow-level features. A unified relation modeling is proposed to remove complex handcrafted layer pattern designs. SuperSBT is further improved by masked image modeling pre-training, integrating temporal modeling, and equipping with dedicated prediction heads. Thus, SuperSBT outperforms the SBT baseline by 4.7%,3.0%, and 4.5% AUC scores in LaSOT, TrackingNet, and GOT-10K. Notably, SuperSBT greatly raises the speed of SBT from 37 FPS to 81 FPS. Extensive experiments show that our method achieves superior results on eight VOT benchmarks.

Sheetal Harris, Jinshuo Liu, Hassan Jalil Hadi et al.

Misinformation can seriously impact society, affecting anything from public opinion to institutional confidence and the political horizon of a state. Fake News (FN) proliferation on online websites and Online Social Networks (OSNs) has increased profusely. Various fact-checking websites include news in English and barely provide information about FN in regional languages. Thus the Urdu FN purveyors cannot be discerned using factchecking portals. SOTA approaches for Fake News Detection (FND) count upon appropriately labelled and large datasets. FND in regional and resource-constrained languages lags due to the lack of limited-sized datasets and legitimate lexical resources. The previous datasets for Urdu FND are limited-sized, domain-restricted, publicly unavailable and not manually verified where the news is translated from English into Urdu. In this paper, we curate and contribute the first largest publicly available dataset for Urdu FND, Ax-to-Grind Urdu, to bridge the identified gaps and limitations of existing Urdu datasets in the literature. It constitutes 10,083 fake and real news on fifteen domains collected from leading and authentic Urdu newspapers and news channel websites in Pakistan and India. FN for the Ax-to-Grind dataset is collected from websites and crowdsourcing. The dataset contains news items in Urdu from the year 2017 to the year 2023. Expert journalists annotated the dataset. We benchmark the dataset with an ensemble model of mBERT,XLNet, and XLM RoBERTa. The selected models are originally trained on multilingual large corpora. The results of the proposed model are based on performance metrics, F1-score, accuracy, precision, recall and MCC value.

Yue Cao, Quansong He, Kaishen Wang et al.

U-like networks have become fundamental frameworks in medical image segmentation through skip connections that bridge high-level semantics and low-level spatial details. Despite their success, conventional skip connections exhibit two key limitations: inter-feature constraints and intra-feature constraints. The inter-feature constraint refers to the static nature of feature fusion in traditional skip connections, where information is transmitted along fixed pathways regardless of feature content. The intra-feature constraint arises from the insufficient modeling of multi-scale feature interactions, thereby hindering the effective aggregation of global contextual information. To overcome these limitations, we propose a novel Dynamic Skip Connection (DSC) block that fundamentally enhances cross-layer connectivity through adaptive mechanisms. The DSC block integrates two complementary components. (1) Test-Time Training (TTT) module. This module addresses the inter-feature constraint by enabling dynamic adaptation of hidden representations during inference, facilitating content-aware feature refinement. (2) Dynamic Multi-Scale Kernel (DMSK) module. To mitigate the intra-feature constraint, this module adaptively selects kernel sizes based on global contextual cues, enhancing the network capacity for multi-scale feature integration. The DSC block is architecture-agnostic and can be seamlessly incorporated into existing U-like network structures. Extensive experiments demonstrate the plug-and-play effectiveness of the proposed DSC block across CNN-based, Transformer-based, hybrid CNN-Transformer, and Mamba-based U-like networks.

Le Yu, Kaishen Wang, Jianlong Xiong et al.

Though Large Vision-Language Models (LVLMs) have achieved remarkable performance across various tasks, they are still prone to hallucinations-generating outputs that are textually plausible but visually ungrounded. While prior approaches generally address this issue through data-centric fine-tuning or innovative decoding strategies, these methods often require substantial resources or task-specific configurations. In this work, we introduce an architecture-level solution, HalluRNN, which enhances model stability through recurrent cross-layer reasoning. Specifically, we propose a novel Dual-Gated Depth Propagation Unit (DG-DPU) module, which is shared across layers and recurrently refines hidden states. This allows for the adaptive propagation of information throughout the model, enforces consistency across layers, and mitigates hallucinations caused by representational drift. By fine-tuning only the DG-DPU module, HalluRNN achieves strong and robust performance across multiple benchmarks.

Yihao Huang, Yue Cao, Tianlin Li et al.

Segment anything model (SAM) has presented impressive objectness identification capability with the idea of prompt learning and a new collected large-scale dataset. Given a prompt (e.g., points, bounding boxes, or masks) and an input image, SAM is able to generate valid segment masks for all objects indicated by the prompts, presenting high generalization across diverse scenarios and being a general method for zero-shot transfer to downstream vision tasks. Nevertheless, it remains unclear whether SAM may introduce errors in certain threatening scenarios. Clarifying this is of significant importance for applications that require robustness, such as autonomous vehicles. In this paper, we aim to study the testing-time robustness of SAM under adversarial scenarios and common corruptions. To this end, we first build a testing-time robustness evaluation benchmark for SAM by integrating existing public datasets. Second, we extend representative adversarial attacks against SAM and study the influence of different prompts on robustness. Third, we study the robustness of SAM under diverse corruption types by evaluating SAM on corrupted datasets with different prompts. With experiments conducted on SA-1B and KITTI datasets, we find that SAM exhibits remarkable robustness against various corruptions, except for blur-related corruption. Furthermore, SAM remains susceptible to adversarial attacks, particularly when subjected to PGD and BIM attacks. We think such a comprehensive study could highlight the importance of the robustness issues of SAM and trigger a series of new tasks for SAM as well as downstream vision tasks.

Yue Cao, Fatemeh H. Fard

Studies show that developers' answers to the mobile app users' feedbacks on app stores can increase the apps' star rating. To help app developers generate answers that are related to the users' issues, recent studies develop models to generate the answers automatically. Aims: The app response generation models use deep neural networks and require training data. Pre-Trained neural language Models (PTM) used in Natural Language Processing (NLP) take advantage of the information they learned from a large corpora in an unsupervised manner, and can reduce the amount of required training data. In this paper, we evaluate PTMs to generate replies to the mobile app user feedbacks. Method: We train a Transformer model from scratch and fine-tune two PTMs to evaluate the generated responses, which are compared to RRGEN, a current app response model. We also evaluate the models with different portions of the training data. Results: The results on a large dataset evaluated by automatic metrics show that PTMs obtain lower scores than the baselines. However, our human evaluation confirms that PTMs can generate more relevant and meaningful responses to the posted feedbacks. Moreover, the performance of PTMs has less drop compared to other models when the amount of training data is reduced to 1/3. Conclusion: PTMs are useful in generating responses to app reviews and are more robust models to the amount of training data provided. However, the prediction time is 19X than RRGEN. This study can provide new avenues for research in adapting the PTMs for analyzing mobile app user feedbacks. Index Terms-mobile app user feedback analysis, neural pre-trained language models, automatic answer generation

Florin C. Ghesu, Bogdan Georgescu, Awais Mansoor et al.

Building accurate and robust artificial intelligence systems for medical image assessment requires not only the research and design of advanced deep learning models but also the creation of large and curated sets of annotated training examples. Constructing such datasets, however, is often very costly -- due to the complex nature of annotation tasks and the high level of expertise required for the interpretation of medical images (e.g., expert radiologists). To counter this limitation, we propose a method for self-supervised learning of rich image features based on contrastive learning and online feature clustering. For this purpose we leverage large training datasets of over 100,000,000 medical images of various modalities, including radiography, computed tomography (CT), magnetic resonance (MR) imaging and ultrasonography. We propose to use these features to guide model training in supervised and hybrid self-supervised/supervised regime on various downstream tasks. We highlight a number of advantages of this strategy on challenging image assessment problems in radiography, CT and MR: 1) Significant increase in accuracy compared to the state-of-the-art (e.g., AUC boost of 3-7% for detection of abnormalities from chest radiography scans and hemorrhage detection on brain CT); 2) Acceleration of model convergence during training by up to 85% compared to using no pretraining (e.g., 83% when training a model for detection of brain metastases in MR scans); 3) Increase in robustness to various image augmentations, such as intensity variations, rotations or scaling reflective of data variation seen in the field.

Muhammad Khalid, Liang Wang, Kezhi Wang et al.

In this paper, to reduce the congestion rate at the city center and increase the quality of experience (QoE) of each user, the framework of long-range autonomous valet parking (LAVP) is presented, where an Autonomous Vehicle (AV) is deployed in the city, which can pick up, drop off users at their required spots, and then drive to the car park out of city center autonomously. In this framework, we aim to minimize the overall distance of the AV, while guarantee all users are served, i.e., picking up, and dropping off users at their required spots through optimizing the path planning of the AV and number of serving time slots. To this end, we first propose a learning based algorithm, which is named as Double-Layer Ant Colony Optimization (DL-ACO) algorithm to solve the above problem in an iterative way. Then, to make the real-time decision, while consider the dynamic environment (i.e., the AV may pick up and drop off users from different locations), we further present a deep reinforcement learning (DRL) based algorithm, which is known as deep Q network (DQN). The experimental results show that the DL-ACO and DQN-based algorithms both achieve the considerable performance.

Yue Cao, Xiaohe Wu, Shuran Qi et al.

The success of deep denoisers on real-world color photographs usually relies on the modeling of sensor noise and in-camera signal processing (ISP) pipeline. Performance drop will inevitably happen when the sensor and ISP pipeline of test images are different from those for training the deep denoisers (i.e., noise discrepancy). In this paper, we present an unpaired learning scheme to adapt a color image denoiser for handling test images with noise discrepancy. We consider a practical training setting, i.e., a pre-trained denoiser, a set of test noisy images, and an unpaired set of clean images. To begin with, the pre-trained denoiser is used to generate the pseudo clean images for the test images. Pseudo-ISP is then suggested to jointly learn the pseudo ISP pipeline and signal-dependent rawRGB noise model using the pairs of test and pseudo clean images. We further apply the learned pseudo ISP and rawRGB noise model to clean color images to synthesize realistic noisy images for denoiser adaption. Pseudo-ISP is effective in synthesizing realistic noisy sRGB images, and improved denoising performance can be achieved by alternating between Pseudo-ISP training and denoiser adaption. Experiments show that our Pseudo-ISP not only can boost simple Gaussian blurring-based denoiser to achieve competitive performance against CBDNet, but also is effective in improving state-of-the-art deep denoisers, e.g., CBDNet and RIDNet.

Florin C. Ghesu, Bogdan Georgescu, Awais Mansoor et al.

The interpretation of medical images is a challenging task, often complicated by the presence of artifacts, occlusions, limited contrast and more. Most notable is the case of chest radiography, where there is a high inter-rater variability in the detection and classification of abnormalities. This is largely due to inconclusive evidence in the data or subjective definitions of disease appearance. An additional example is the classification of anatomical views based on 2D Ultrasound images. Often, the anatomical context captured in a frame is not sufficient to recognize the underlying anatomy. Current machine learning solutions for these problems are typically limited to providing probabilistic predictions, relying on the capacity of underlying models to adapt to limited information and the high degree of label noise. In practice, however, this leads to overconfident systems with poor generalization on unseen data. To account for this, we propose a system that learns not only the probabilistic estimate for classification, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output. We argue that this approach is essential to account for the inherent ambiguity characteristic of medical images from different radiologic exams including computed radiography, ultrasonography and magnetic resonance imaging. In our experiments we demonstrate that sample rejection based on the predicted uncertainty can significantly improve the ROC-AUC for various tasks, e.g., by 8% to 0.91 with an expected rejection rate of under 25% for the classification of different abnormalities in chest radiographs. In addition, we show that using uncertainty-driven bootstrapping to filter the training data, one can achieve a significant increase in robustness and accuracy.

Abdelrahman Abdelhamed, Mahmoud Afifi, Radu Timofte et al.

This paper reviews the NTIRE 2020 challenge on real image denoising with focus on the newly introduced dataset, the proposed methods and their results. The challenge is a new version of the previous NTIRE 2019 challenge on real image denoising that was based on the SIDD benchmark. This challenge is based on a newly collected validation and testing image datasets, and hence, named SIDD+. This challenge has two tracks for quantitatively evaluating image denoising performance in (1) the Bayer-pattern rawRGB and (2) the standard RGB (sRGB) color spaces. Each track ~250 registered participants. A total of 22 teams, proposing 24 methods, competed in the final phase of the challenge. The proposed methods by the participating teams represent the current state-of-the-art performance in image denoising targeting real noisy images. The newly collected SIDD+ datasets are publicly available at: https://bit.ly/siddplus_data.

Yue Cao, Tianlong Chen, Zhangyang Wang et al.

Learning to optimize has emerged as a powerful framework for various optimization and machine learning tasks. Current such "meta-optimizers" often learn in the space of continuous optimization algorithms that are point-based and uncertainty-unaware. To overcome the limitations, we propose a meta-optimizer that learns in the algorithmic space of both point-based and population-based optimization algorithms. The meta-optimizer targets at a meta-loss function consisting of both cumulative regret and entropy. Specifically, we learn and interpret the update formula through a population of LSTMs embedded with sample- and feature-level attentions. Meanwhile, we estimate the posterior directly over the global optimum and use an uncertainty measure to help guide the learning process. Empirical results over non-convex test functions and the protein-docking application demonstrate that this new meta-optimizer outperforms existing competitors.

Bin Liu, Yue Cao, Mingsheng Long et al.

Deep hashing establishes efficient and effective image retrieval by end-to-end learning of deep representations and hash codes from similarity data. We present a compact coding solution, focusing on deep learning to quantization approach that has shown superior performance over hashing solutions for similarity retrieval. We propose Deep Triplet Quantization (DTQ), a novel approach to learning deep quantization models from the similarity triplets. To enable more effective triplet training, we design a new triplet selection approach, Group Hard, that randomly selects hard triplets in each image group. To generate compact binary codes, we further apply a triplet quantization with weak orthogonality during triplet training. The quantization loss reduces the codebook redundancy and enhances the quantizability of deep representations through back-propagation. Extensive experiments demonstrate that DTQ can generate high-quality and compact binary codes, which yields state-of-the-art image retrieval performance on three benchmark datasets, NUS-WIDE, CIFAR-10, and MS-COCO.

Weifeng Gao, Zhiwei Zhao, Geyong Min et al.

Body Sensor Network (BSN) is a typical Internet-of-Things (IoT) application for personalized health care. It consists of economically powered, wireless and implanted medical monitoring sensor nodes, which are designed to continually collect the medical information of the target patients. Multichannel is often used in BSNs to reduce the spectrum competition of the tremendous sensor nodes and the problem of channel assignment has attracted much research attention. The health sensing data in BSNs is often required to be delivered to a sink node (or server) before a certain deadline for real time monitoring or health emergency alarm. Therefore, deadline is of significant importance for multichannel allocation and scheduling. The existing works, though designed to meet the deadline, often overlook the impact of the unreliable wireless links. As a result, the health sensing data can still be overdue because of the scheduled lossy links. Besides, potential collisions in the schedules also incur considerable delay in delivering the sensing data. In this paper, we propose a novel deadline- driven Link quality Aware Channel Assignment scheme (LACA), where link quality, deadlines and collisions are jointly considered. LACA prioritizes links with urgent deadlines and heavy collisions. Besides, LACA allows the exploition of the spare slots for retransmissions on lossy links, which can further reduce the retransmission delay. Extensive simulation experiments show that compared to the existing approaches, LACA can better utilize the wireless spectrum and achieve higher packet delivery ratio before the deadline.

Yue Cao, Mingsheng Long, Jianmin Wang et al.

Hashing is widely applied to approximate nearest neighbor search for large-scale multimodal retrieval with storage and computation efficiency. Cross-modal hashing improves the quality of hash coding by exploiting semantic correlations across different modalities. Existing cross-modal hashing methods first transform data into low-dimensional feature vectors, and then generate binary codes by another separate quantization step. However, suboptimal hash codes may be generated since the quantization error is not explicitly minimized and the feature representation is not jointly optimized with the binary codes. This paper presents a Correlation Hashing Network (CHN) approach to cross-modal hashing, which jointly learns good data representation tailored to hash coding and formally controls the quantization error. The proposed CHN is a hybrid deep architecture that constitutes a convolutional neural network for learning good image representations, a multilayer perception for learning good text representations, two hashing layers for generating compact binary codes, and a structured max-margin loss that integrates all things together to enable learning similarity-preserving and high-quality hash codes. Extensive empirical study shows that CHN yields state of the art cross-modal retrieval performance on standard benchmarks.